Depression is a prevalent, chronic, and recurrent psychiatric disorder with significant associated morbidity, mortality, and economic costs. Despite the efficacy of antidepressant medications, treatment response time can take as long as 6-8 weeks and 20-35% of patients fail to respond adequately. Innovative treatments that accelerate the onset of antidepressant action and increase response and remission rates are required to reduce depression burden. Partial sleep deprivation (PSD) is a safe non-pharmacological treatment with demonstrated antidepressant efficacy that could be combined with antidepressant medication to augment treatment response. Our pilot data suggest that mood improvement occurs with a modest amount of PSD repeated over weeks of treatment. Our data further suggest that non-responders to fluoxetine treatment have shorter REM latency and more REM sleep than non-responders, particularly in the last few hours of the night. The objectives of the proposed study are to evaluate the efficacy and safety of adjunctive early and late PSD compared to no sleep deprivation (NSD) for augmenting the effects of 8 weeks of fluoxetine 20-40 mg treatment and to examine the underlying sleep mechanisms of treatment response. One hundred and twenty symptomatic but unmedicated depressed patients will receive fluoxetine 20-40 mg for 8 weeks and be randomly assigned to one of three time in bed (TIB) conditions: (1) no sleep deprivation (NSD, 2300 to 0700 hours); (2) early PSD (E-PSD, 0100 to 0700 hours); or (3) late PSD (L-PSD, 2300 to 0500 hours). Participants will undergo three pre-treatment in-laboratory nights (adaptation, baseline, TIB condition), 12 consecutive at- home nights with fluoxetine and the assigned TIB condition, and 2 post-treatment in-laboratory nights (TIB condition, NSD). Outcome measures will include self- and clinician-rated mood, pre-and post-TIB condition sleep, and neurobehavioral testing to evaluate the daytime consequences of the PSD procedure. The specific aims of the study are: (1) to compare the acute mood effects of 2 hours of early-night partial sleep deprivation (E-PSD), 2 hours of late-night partial sleep deprivation (L-PSD), and no sleep deprivation (NSD) in patients with depression; (2) to evaluate whether a treatment combining fluoxetine 20-40 mg with two weeks of repeated 1-hour PSD (E-PSD or L-PSD) is more efficacious than fluoxetine 20-40 mg plus NSD in patients with depression; (3) to determine which EEG measures define treatment response; and (4) to evaluate neurobehavioral functioning before PSD, after 1 and 14 consecutive nights of PSD, and at 8-week clinical follow-up. The long-term objective of this research is to develop an effective and safe adjunctive non- pharmacological intervention that can be implemented in the clinical setting to improve the clinical course for patients with depression. PUBLIC HEALTH RELEVANCE: Depression is a prevalent, chronic, and recurrent psychiatric disorder with significant associated morbidity, mortality, and economic costs. Antidepressant medications, while effective, can take 6-8 weeks to work and up to one-third patients fail to have an adequate treatment response. Efficacious and safe adjunctive treatments that accelerate antidepressant efficacy and improve overall response rates are needed. The proposed study will test the benefit and safety of an adjunctive PSD intervention that can be easily implemented in the clinical setting and that may ultimately substantially improve the clinical course for patients with depression.